S1950
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
Material/Methods: Ten patients with cervical to thoracic spine metastases were replanned with 20 Gy in one fraction using the Eclipse Treatment Planning System (version 18.1, Varian Medical Systems, Palo Alto, CA, USA). The VMATDLS plan used two arcs with static collimator angles, and three DLS were generated to control dose fall-off. Five medical physicists independently created DLS to evaluate the manual processing time. The RADSBRT-NTO plan utilized a single arc, incorporating dynamic collimator rotation with seven static angle modulated ports and applying the SBRT NTO. All plans adhered to the RTOG 0631 criteria and were normalized to the prescribed dose covering 90% of the PTV. Target coverage, spinal cord sparing, and plan quality indices, including the Paddick conformity index (PCI) and gradient index (GI), were assessed. Treatment delivery time, plan optimization time, and total monitor units (MUs) were recorded to assess planning efficiency. Results: The RADSBRT-NTO plan with only a single arc achieved comparable dosimetry quality to VMATDLS. No significant differences were observed in target coverage, spinal cord sparing, PCI, or GI (Table 1). RADSBRT-NTO significantly improved planning efficiency (p < 0.001), resulting in a 38% reduction in treatment delivery time and total MUs (Table 2). Furthermore, optimization time was shortened by 81%,and the time for manual DLS creation (1.1 ± 0.3 min) was nearly equivalent to the optimization time for RADSBRT-NTO. This DLS-free workflow resulted in an overall 84% reduction in total planning time.
convenient DLS-free and highly efficient SSRS workflow for spine metastasis. References: [1] Varian Medical Systems, Eclipse TPS 18.1 New Features Workbook. 2024. Keywords: RapidArc Dynamic, Stereotactic spine radiosurgery Digital Poster 3786 Development of a model for Gradient Index prediction for Linac-based Stereotactic Radiotherapy for brain metastases Lorenzo Marini 1 , Andrea Bruschi 2 , Alessandro Ghirelli 2 , Silvia Pini 2 , Claudia Poggiali 2 , Edoardo Salmeri 3 , Roberto Pellegrini 3 , Serenella Russo 2 1 Department of Experimental and Clinical Biomedical Sciences “M. Serio”, ”, University of Florence, Florence, Florence, Italy. 2 Medical Physics Unit Florence-Empoli, Department of Hospitals Network, Azienda USL Toscana Centro, Florence, Italy. 3 Clinical Research – Medical Affairs, Elekta AB, Stockholm, Sweden Purpose/Objective: VMAT planning for cranial Radiosurgery (SRS) or Hypofractionated Stereotactic Radiotherapy (HFSRT) treatments of patients with single or multiple brain lesions using single isocenter demonstrated that highly conformal dose distributions can be achieved with optimal target coverage and dose fall-off in the high dose area. To predict the expected dose gradient as a function of the target volume size could be essential for standardizing the planning phase. Material/Methods: Single or multiple brain metastases SRS and HFSRT treatments for 66 patients performed on an Elekta Versa-HD Linac were analyzed. VMAT SRS/HFSRT Planning was automated by using a robust template- based optimization in Elekta MONACO TPS with 5 non- coplanar arcs for a prescription dose of 20 Gy in single fraction or 27 Gy/3 fractions or 30 Gy /5 fractions for HFSRT. For each plan, geometric and dosimetric parameters were collected, including Target Volume, Maximum Dose, Prescription Isodose Level, Paddick Conformity Index and Gradient Index (GI). A non-linear inverse power-law model was tested to describe the GI–PTV relationship: GI = A/PTVB, where PTV is the target volume. Model fitting was performed with non- linear least squares and fit accuracy was evaluated with Mean Squared Error (MSE), Mean Absolute Error (MAE), and R-Squared (R ² ) metrics. Results: A dataset of 144 cranial plans was retrospectively analyzed. Median tumor volume was 0.59 cc (range 0.18–33.77 cc), median maximum dose was 25.9 Gy (range 22.9–38.6 Gy), and the prescription isodose
Conclusion: SBRT NTO–based RAD planning achieved dosimetric quality comparable to ring-based VMAT using a single arc. SBRT NTO–based RAD planning enables a
Made with FlippingBook - Share PDF online